Multiple models indicate that contraceptive use for prevention of unplanned pregnancy is the most cost- effective way to prevent mother to child transmission of HIV. Though multiple contraceptive methods are available to HIV-positive women, method discontinuation and unintended pregnancy are common events. Further, data have emerged suggesting systemic hormonal methods may accelerate HIV disease progression and increase transmission to male partners. In South Africa, HIV prevalence among reproductive-aged women is high (17%) and contraceptive use rates over a 12 month period are low (29/100 woman-years). The levonorgestrel intrauterine device (LNG IUD) is a long-acting, highly effective contraceptive method with largely localized hormonal effects, 10-log lower serum progestin concentrations than available systemic hormonal methods, and high uptake and continuation rates in diverse settings among HIV-negative women. Preliminary data from 277 HIV-positive women in Cape Town indicate that, though 37% were aware of IUDs, 75% were willing to consider the IUD for contraception following information about IUDs. Expanding method choice may increase coverage and a low-cost LNG IUD is likely to become available in the next few years. Thus, we propose a randomized clinical trial in Cape Town, South Africa with 320 HIV-positive ART-na?ve (CD4>500) women enrolled in a 1:1 ratio to Intervention (LNG IUD) and Control (copper T380 IUD) arms and followed at 3, 6, 9, 12, 18, and 24 months to determine comparative safety and acceptability of the LNG IUD. Participants and outcome assessors will be blinded to treatment arm. We will compare the study arms to address three specific aims: 1) To assess LNG IUD safety through comparative impact of low exogenous progestin exposure on HIV progression markers, including plasma viral load (VL) and CD4 change, and eligibility for ART initiation, or other health status indicators, including sexually-transmitted infection incidence; 2) To measure LNG IUD safety with respect to genital HIV-1 RNA shedding, a surrogate for transmission, measured with genital tract VL; and 3) To measure LNG IUD acceptability by comparing IUD continuation, pregnancy, and expulsion rates between study arms. We will also measure bleeding changes and method satisfaction using qualitative and quantitative measures for participants in each arm, a novel design aspect. This study will inform international medical guidelines to definitively establish LNG IUD safety with regard to HIV progression and transmission, which is of great importance as systemic hormonal methods may become contraindicated due to adverse effects on HIV transmission and progression. The potential impact through subsequent introduction of this highly effective and long-acting contraceptive would be significant reduction of unplanned pregnancy and mother-to-child transmission of HIV, an NIH Office of AIDS Research priority for 2011, and provision of non- contraceptive benefits, such as reduction of anemia, to HIV-positive women in Sub-Saharan Africa.